Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.
Arena text preference / chinesehuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,498.12 rating
Confidence interval: 1,490.36 – 1,505.87
6888 votes
Category
chinese
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,463.87 rating
Confidence interval: 1,459.18 – 1,468.57
27540 votes
Category
coding
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,402.23 rating
Confidence interval: 1,396.72 – 1,407.73
17395 votes
Category
creative_writing
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,443.20 rating
Confidence interval: 1,439.21 – 1,447.19
42806 votes
Category
english
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,433.07 rating
Confidence interval: 1,429.01 – 1,437.14
71392 votes
Category
exclude_ties
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,461.44 rating
Confidence interval: 1,454.89 – 1,467.99
10502 votes
Category
expert
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,462.61 rating
Confidence interval: 1,450.41 – 1,474.82
3179 votes
Category
french
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,443.85 rating
Confidence interval: 1,427.68 – 1,460.02
1516 votes
Category
german
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,447.63 rating
Confidence interval: 1,444.02 – 1,451.24
64295 votes
Category
hard_prompts
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Qwen3.5 397B-A17B
1,450.23 rating
Confidence interval: 1,445.60 – 1,454.85
28678 votes
Category
hard_prompts_english
Evaluated configuration
Qwen3.5-397B-A17B
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,398.15 rating
Confidence interval: 1,393.01 – 1,403.29
21199 votes
Category
industry_entertainment_and_sports_and_media
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,440.36 rating
Confidence interval: 1,432.81 – 1,447.92
7182 votes
Category
industry_legal_and_government
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,456.55 rating
Confidence interval: 1,451.01 – 1,462.08
15315 votes
Category
industry_life_and_physical_and_social_science
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,453.40 rating
Confidence interval: 1,444.78 – 1,462.01
5472 votes
Category
industry_mathematical
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,454.23 rating
Confidence interval: 1,446.45 – 1,462.00
6965 votes
Category
industry_medicine_and_healthcare
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,463.19 rating
Confidence interval: 1,458.97 – 1,467.42
37377 votes
Category
industry_software_and_it_services
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,415.18 rating
Confidence interval: 1,410.28 – 1,420.07
23093 votes
Category
industry_writing_and_literature_and_language
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,423.13 rating
Confidence interval: 1,418.71 – 1,427.54
31992 votes
Category
instruction_following
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,428.69 rating
Confidence interval: 1,409.33 – 1,448.06
1136 votes
Category
japanese
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,382.68 rating
Confidence interval: 1,366.39 – 1,398.97
1570 votes
Category
korean
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,441.66 rating
Confidence interval: 1,437.34 – 1,445.97
41250 votes
Category
longer_query
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,453.59 rating
Confidence interval: 1,444.97 – 1,462.21
5146 votes
Category
math
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,446.23 rating
Confidence interval: 1,440.55 – 1,451.91
15114 votes
Category
multi_turn
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,429.70 rating
Confidence interval: 1,425.99 – 1,433.41
52251 votes
Category
non_english
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,437.72 rating
Confidence interval: 1,434.65 – 1,440.79
93567 votes
Category
overall
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,448.51 rating
Confidence interval: 1,433.99 – 1,463.04
1665 votes
Category
polish
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,429.31 rating
Confidence interval: 1,422.98 – 1,435.64
10715 votes
Category
russian
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
Qwen3.5 397B-A17B
1,440.53 rating
Confidence interval: 1,428.65 – 1,452.41
2935 votes
Category
spanish
Evaluated configuration
qwen3.5-397b-a17b
Evaluation type
human preference
Run it locally
Inference · memory estimate
Qwen3.5 397B-A17B
751.4 GiB checkpoint tensor payload · 403.40 B stored tensor elements · BF16/F32
Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.
API cost
USD / month
Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.